[jbraithwaite]

Hi there,

I'm a mechanical engineering student and a bit stuck specifying a weld size (bead throat) for a design I'm working on.

The pieces to be welded are 2mm thick 6082 extrusions.

Strength calculations I've done indicate that a 1mm throat would suffice (using 5356 filler), however I have no idea what a reasonable minimum weld size is.

I have been taught that the size should be between approx. 0.6 and 0.8 times the material thickness.

I've seen a couple of examples of 2mm aluminium welded with quite thick welds, so it seems it is possible to exceed the 0.8*t rule of thumb without burning a hole in the extrusions...

I'm suggesting TIG welds as it seems to be the preferred method of welding thin aluminium (less heat penetration than MIG).

My questions are:
1) What is the minimum weld size (bead throat) that is achievable with TIG?
2) Is it safe to exceed the rule of thumb I mentioned above?

Thanks
James

[Welder2008]

Bigger is ok, unless there's a specific reason it can't be like a larger weld interfering with fit-up of a secondary part, throwing something out of balance, or called out on the spec that it can not exceed a certain size. It my experience, the LEG size of a fillet weld should be 1t (one material thickness) which would put the effective throat size in the approximate 0.6-0.8t range like you said. As an engineer you should understand that it's all about load ratings (static, dynamic, cyclic), strength in base material, strength in weld material, and safety factors.... Yes, you can go larger on your weld size, long as it does not pose a problem down the line. Depending on circumstances mentioned, I would just call out on the print for your project a weld with a leg size equal to your material thickness because weld sizes on manufacturing prints are never called out as the effective throat size. The minimum achievable weld size capable with tig is solely dependant upon the person with the torch in their hand. I used to weld aircraft parts with a material thickness of 0.010" (butt weld) and spec called out that the reinforcement on either side could not exceed 1t (0.030" total weld thickness) and it was checked with dropping a machined ball bearing down the tube and sliding a machined ring around the outside of the tubing which was 0.250"-1.500" diameter.

[Otto Nobedder]

Welcome aboard, James,

Your name sounds very familiar to me for some reason. I suppose it's not unique, though, so may be coincidence.

The hazards of excessive weld size include overheating the base metal, creating too wide a HAZ, making a potential weakness. On the other hand, when you specify a leg dimension of 1t, with the expectation of 0.6-0.8t throat, this should be an absolute minimum with no negative tolerance. A positive tolerance is required, as no welder is perfect. I'd suggest (from experience only; no documentation to back it) a leg dimension of 1t +0.5t -0t. This, if met, should guarantee your required throat thickness. (Note, as material thickness increases, 1.5t becomes quite excessive.)

There are two weak points in an aluminum GTAW weld that the inspectors should pay attention to on the initial VT. Does the start penetrate the fillet completely, or just "bridge" at the start point? That's a potential stress riser.
Is the termination filled, or did the welder leave a crater? A termination crater should be a VT fail. It's not only a stress riser, but with many aluminum allows it will cold-crack before the part sees external stress.

If the welder produced both problems, the whole weld is suspect for LOF in the root.

Disclaimer... I'm not an engineer. I've just been on the end of a torch for a lot of years, and I'm relating experience. Your mileage may vary.

Steve S

[jbraithwaite]

Thank you Welder2008 and Otto.

Both of your posts have been really interesting and informative. With most of my career thus far spent focusing on theory and in an academic environment it's really valuable to get advice from people with a lot of practical experience.

My main concern with a larger weld size was overheating the extrusions as they are so thin (and aluminium). I'm not too worried about the HAZ as I have a reasonable safety factor and the load isn't that high. I also don't expect fatigue to be a problem, so I'm not too concerned about stress raisers etc.

So would I be safe to spec a 2mm leg size? The design is for an assignment so I will receive a horrified look from my lecturer if I spec a size smaller than what is possible to weld.

Otto, I'm afraid I can't say your name rings a bell. It's most likely just coincidence. I attend the University of Cape Town in South Africa.

Thanks again.
James

[Otto Nobedder]

I would think the 2mm leg size would be sufficient, as long as the tolerance is solely to the positive side (WPS range of 2.0 to 2.5mm, for example).

And, yes, I can do a weld that small in aluminum, even with the limited resources I have at work. With a modern inverter machine with adjustable frequency and a 1/16" lanthanated tungsten, this weld should be trainable for almost anyone with a year's experience welding aluminum.

And, I agree, considering where you're studying, I doubt we've met, unless I've encountered you on facebook.

Steve S

[Oscar]

yes, I can do a weld that small in aluminum

Yea James, you ought'a know betta' that Otto Nobedder can produce a good weld that size!